-
Learning an augmentation strategy for sparse datasets
-
A dual-aligned knowledge self-distillation framework for visible-infrared cross-modal person re-identification
• Dual alignment knowledge self-distillation to better capture modality-invariant/specific features for VI-ReID • Temperature-modulated alignment and confidence-based selective masking to enhance model reliability. • CutSwap augmentation to improve model robustness against intra-class variations and modality discrepancies. • State-of-the-art performance on…
-
Automatic 3D Object Segmentation in Multiple Views using Volumetric Graph-Cuts
We propose an algorithm for automatically obtaining a segmentation of a rigid object in a sequence of images that are calibrated for camera pose and intrinsic parameters. Until recently, the best segmentation results have been obtained by interactive methods that…
-
DPN: Density-Based Probability VANET Caching Method Leveraging with NDN Concept
Vehicular Ad Hoc Network (VANET) is expected to have broad applications, such as traffic warnings and accident rescue. However, the highly dynamic nature of VANET topology challenges conventional Mobile Ad-hoc Network (MANET) data retrieval mecha-nisms, causing significant performance degradation. To…
-
DPNVC: A novel density-based probability VANET caching framework built upon the NDN
-
Fully-Autonomous, Vision-based Traffic Signal Control: From Simulation to Reality
Ineffective traffic signal control is one of the major causes of congestion in urban road networks. Dynamically changing traffic conditions and live traffic state estimation are fundamental challenges that limit the ability of the existing signal infrastructure in rendering individualized…
-
Multi-person 3D pose estimation from unlabelled data
Abstract Its numerous applications make multi-human 3D pose estimation a remarkably impactful area of research. Nevertheless, it presents several challenges, especially when approached using multiple views and regular RGB cameras as the only input. First, each person must be uniquely…
-
Reconstructing relief surfaces
-
WITHDRAWN: Video-based, real-time multi-view stereo
-
Reconstructing Relief Surfaces
This paper generalizes Markov Random Field (MRF) stereo methods to the generation of surface relief (height) fields rather than disparity or depth maps. This generalization enables the reconstruction of complete object models using the same algorithms that have been previously…
